The focus of this renewal application will be on the development of single particle cryo-EM structure determination methods that incorporate validation, assessment of alignment errors, and cross-validation of observed conformational variability. We will concentrate on three specific areas: (1) establishment of a well- defined goal function for structure determination, (2) structure refinement methods that incorporate validation of the outcome, and (3) quantitative analysis of conformational variability cross-validated by X-ray model-based simulations. In (1), we will establish a goal function (a single-valued function of the 3D map and/or projection data) that would have a global extremum for the correct structure and which can be related to intuitive notion of resolution of the map. A goal function with such properties would make possible a critical evaluation of the ability of existing structure determination methods to deliver optimal structures and establish a theoretical basis for unification and rationalization of 3D-EM single particle structure determination methodology. In (2), we will introduce novel quantitative single particle cryo-EM methodology, based on jackknife-d resampling, designed to yield an estimate of alignment errors and eliminate reference bias that results in artifactual features that are indistinguishable from genuine ones in the absence of external standards. Incorporation of these developments into cryo-EM structure determination algorithms will make possible objective validation of a refined 3D map. They will provide, for the first time, a simple but robust way to eliminate artifacts and will thus increase the level of confidence in cryo-EM results. In (3), we will use our projection data resampling methodology to calculate (directly from the data) eigenvectors characterizing the conformational variability of a structure. We will then develop a deconvolution algorithm that will use this eigenvector information to eliminate from a 3D map the blurring caused by residual alignment errors. We will also use the eigenanalysis to characterize local mobility of a macromolecule and bridge the gap between experimental cryo-EM structure determination and simulations of conformational variability based on physical models. By cross-validating our methodology with the results of molecular dynamics simulations we will provide a novel tool for analyzing the energy landscape of large macromolecules. Rather than incremental improvements, the methods we propose to develop will put single particle cryo-EM analysis on a new path towards full reliability of the results, eliminating the uncertainty that currently hinder fulfillment of cryo-EM's full potential. To assure multi-platform portability and immediate dissemination, these new methods will be implemented within the SPARX image processing package.